BACKGROUND: Treatment of intra-abdominal infections (IAI) has important health economic components including hospital stay and the cost and administration of pharmaceuticals. METHODS: From the hospital s perspective, time-to-discharge (TTD) was evaluated for 258 efficacy-evaluable (EE), IAI patients who received piperacillin/tazobactam (pip/tazo, 4g/500 mg) or imipenem/cilastatin (imip/cil, 1g/1g) q8 hr by slow infusion in a randomized, double blind study. Baseline severity was evaluated using the APACHE II scoring system. Clinical and bacteriological efficacy was similar across treatment groups. A Cox proportional hazards analysis was used to estimate TTD adjusting for APACHE II scores. RESULTS: Baseline APACHE II scores were higher in the pip/tazo group than the imip/cil group for the intent-to-treat population (p = 0.04) and EE population (p = 0.07). The pip/tazo group had a 23% lower instantaneous rate of hospital discharge (IRHD) compared to the imip/cil group (relative odds = 0.77; p=0.04). Severity adjustment using the APACHE II indicated that treatment-related differences in IRHD were not different (p=0.12). APACHE II independently predicted TTD; a 1-point increase in APACHE II score was associated with a 6% lower IRHD (relative odds = 0.94; p=0.001). Pip/tazo patients were treated for IAI for an average of 7.6 days compared to 7.1 days for imip/cil patients. Combining daily average wholesale price (imip/cil $169.89 v. pip/tazo $64.26) and equivalent administration costs resulted in a saving of $718 per course of therapy. CONCLUSIONS: Treatment with pip/tazo is equivalent to imip/cil in terms of efficacy, duration of treatment and APACHE II-adjusted hospital length-of-stay for IAI, but cost saving due to lower acquisition costs.